Tailored substrates for studies of attached cell cultureM. Mrksich
Department of Chemistry, The University of Chicago, Chicago (Illinois 60637, USA), Fax +1 773 702 0805,e-mail: [email protected]
Received 14 November 1997; received after revision 10 March 1998; accepted 10 March 1998
Abstract. Substrates for studies of the interactions of gold substrate are a class of well-ordered substratesattached cells with extracellular matrix components are and provide a convenient method for tailoring sub-often prepared by allowing a protein to adsorb from strates with ligands, proteins and other groups. Meth-
ods that can pattern the monolayers provide a generalsolution onto a glass or polystyrene substrate. Thisstrategy to create substrates that control the size, shapemethod is simple and effective for many studies, but itand spacing of attached cells. This review illustratescan fail in cases that require rigorous control over therecent work that has used these methods of surfacestructure and composition of adsorbed protein. Self-as-chemistry to create tailored substrates for studies insembled monolayers formed by the spontaneous order-cell biology.ing of terminally functionalized alkanethiols onto a
The growth, differentiation and metabolism of cells areinfluenced by a multitude of signals present in theimmediate environment. Among the most importantsignals are the ligands dissolved in the media and thosepresent on the insoluble extracellular matrix surround-ing cells. Studies of the effects of soluble signals oncellular behavior have utilized defined media and havebeen aided by synthetic methods that can prepare bothnatural and nonnatural peptides and ligands. Studiesof the effects of immobilized ligands, by comparison,have been limited by a lack of convenient experimentalmethods for controlling the composition and structureof substrates for attached cell culture. The most com-mon substrates – those prepared by allowing matrixproteins to adsorb to glass or polystyrene substrates –have the limitation that it is difficult to control andcharacterize the densities and structures of immobilizedproteins.
This limitation has created a need for substrates thathave several characteristics: (i) the adsorbed layer ofprotein is compositionally pure in that only selectproteins, peptides and ligands are present; (ii) the pre-sentation of immobilized proteins is homogeneous inthat they are each adsorbed in a single orientation andconformation; (iii) the density of immobilized proteinscan be controlled and measured; (iv) the immobilizedproteins are stable and persistent in that the activity isnot lost due to denaturation, exchange with proteins inthe medium or degradation by cellular proteases.These substrates will find even wider application withthe introduction of strategies that can change the pat-tern of ligands presented to a cell.The recent development of synthetically flexible surfacechemistries now makes possible the preparation of tai-lored substrates for many studies of the behaviour ofattached cells. This review begins with a summary ofthe advantages and limitations of current methods for
654 M. Mrksich Tailored substrates for studies of attached cell culture
preparing substrates that present adsorbed protein. Thereview then introduces methods from organic surfacescience that can prepare structurally well-defined sub-strates, and gives examples of the use of these substratesfor studies involving attached cell culture. The discus-sion is then extended to include methods from micro-fabrication and lithography that can pattern thestructure of substrates, and the use of these substratesto control the shapes, sizes and positions of attachedcells. The review concludes with a survey of emergingtechniques that can create substrates with dynamic con-trol over structure and properties, and the utility ofthese substrates for studies of cell-substrate interactions.
Substrates with adsorbed protein
The most important method for preparing substratesfor studies involving attached cell culture is to allowproteins or antibodies to adsorb from solution ontoglass or plastic substrates. This method is experimen-tally convenient and general, and it provides substratesthat in many respects resemble the in vivo environmentof cells. Commercial sources of these matrix prepara-tions and even precoated substrates are commonlyavailable.The limitations inherent in this method are all related tothe complex mechanisms underlying protein adsorption(for reviews, see refs 1–3). There is an extensive litera-ture in experimental studies of protein adsorption at thesolid-liquid interface: many of these studies have notdirectly addressed the role of adsorbed protein in sub-strates for attached cell culture, but these studies doprovide general guidelines that are relevant to under-standing the preparation of substrates for cell culture.In solution, most proteins are folded into a single,discrete conformation. But on adsorption to a solidinterface, a protein can assume a heterogeneous popula-tion of structures. Figure 1 summarizes the course ofevents that can follow the initial adsorption of protein(a). These events include lateral diffusion of protein (b)and denaturation of the protein (d) at the interface.
Each of these different forms of protein can desorbfrom the interface (a, c, e), but usually the proteinsremain irreversibly adsorbed to the interface (f), or theycan exchange with soluble protein (g). Even in the nearideal case – that of a single, conformationally stableprotein allowed to adsorb to a structurally homoge-neous substrate – the resulting layer of adsorbedprotein is often heterogeneous in structure. Because thebiological activity of a protein is determined by itsstructure, the properties of these substrates are difficultto control.The small amount of protein that is present in anadsorbed layer makes it extremely difficult to character-ize the changes in structure that follow denaturation orreorientation at a substrate. While physical and spectro-scopic techniques do not yet make routine the determi-nation of molecular structure of proteins adsorbed atinterfaces, several indirect strategies have been used toinfer structural information. In studies of the enzymeribonuclease A adsorbed to mica substrates, for exam-ple, Lee and Belfort showed that the activity of theenzyme increased over a period of 24 h [4]. The changein activity was ascribed to a change in the orientation ofenzyme at the interface. Middlaugh and co-workersused a combination of calorimetry and fluorescencespectroscopy to show that, for several different proteinsand substrates, adsorption destabilized the native struc-ture and increased denaturation of proteins [5]. Theadsorption of protein can also change dramatically withsubtle changes in the structure of the protein. Ramsdenand co-workers, for example, showed that the initialrates of adsorption onto a silicon oxide surface of twocytochrome b5 fragments (that differed by switchingtwo residues in the primary sequence) varied by over10-fold [6].Changes in the conditions present during adsorption –including changes in ionic composition, temperatureand pH – can also have dramatic effects on the struc-ture of adsorbed protein. Tiberg and co-workersshowed that the density of b-casein adsorbed to hydro-phobic silica increased when the dissolved salt waschanged from Na+ to Mg+2 to Ca+2 [7]. The structureof the protein layer also depends on the concentrationof protein in the solution from which it adsorbs. Studiesof the adsorption of fibrinogen to a hydrophobic mono-layer showed that the final density of protein increasedby fivefold when the concentration of protein was in-creased from 3 to 280 mg/ml [8]. A kinetic analysis ofthese data was consistent with a mechanism that in-volved unfolding of the protein after adsorption. Forlow concentrations of protein, the initial rate for ad-sorption is slower and is followed by denaturation. Forhigher concentrations, the surface is rapidly saturatedwith protein, and denaturation is prevented. This effectmakes it difficult to rigorously control the density of
Figure 1. Scheme illustrating the complexities associated with thepreparation of substrates by protein adsorption. Several eventslead to a structurally heterogeneous layer of adsorbed protein: seetext for an explanation.
CMLS, Cell. Mol. Life Sci. Vol. 54, 1998 655Review Article
adsorbed protein. For these reasons, matrix-coated sub-strates must be prepared under exactly the same condi-tions to insure that the structure of adsorbed protein –and hence the biological properties of the substrate –remain constant.The ease and generality with which protein-coated sub-strates can be prepared make them an excellent choicefor many studies of cell-matrix interactions. The pur-pose of the preceding section of the review was toemphasize the limitations of this methodology for thoseapplications that require rigorous control over thestructure of the substrate: for example those that re-quire a quantitative analysis of the relationship betweendensity of immobilized ligand and cellular structure.The use of structurally tailored substrates provides analternate strategy that avoids these limitations. The nextsection introduces a class of structurally tailored sub-strates that allows rigorous control over the presenta-tion of protein and ligands. These substrates have beenimportant in a number of studies in biointerfacial sci-ence, and their utility is growing rapidly.
Structurally well defined substrates
This section will present an overview of the use oforganic thin film substrates for attached cell culture.This section will focus on self-assembled monolayers ofalkanethiolates on gold, because this class of monolayeris synthetically the most versatile (for reviews, see refs 9,10]. Other classes of structurally well defined interfacesinclude crystals [11], metals [12] and supported layers oflipids [13]. While these interfaces make excellent sub-strates for many studies of protein adsorption and cellattachment, they have the primary limitations that theinterfacial structure cannot be tailored easily throughsynthesis and they can have poor structural stability.
Self-assembled monolayers
Self-assembled monolayers (SAMs) of alkanethiolateson gold form upon the adsorption of long-chain alka-nethiols, RSH [R=X(CH2)n, n=11–18] from solutionto a gold surface. For concentrations of alkanethiolnear 1 mM in ethanol, the monolayer assembles in aperiod of 1 to 5 h. The structure of these monolayers iswell established (fig. 2) [14, 15]. The sulphur atomscoordinate to the threefold sites of the gold(111) surfaceto give a close-packed array of alkyl chains. Thesechains are trans-extended and tilted approximately 30 °,and present the terminal functional group X at thesurface; these exposed groups determine the propertiesof the interface. Even alkanethiols that are substitutedwith complex groups assemble into well-ordered mono-layers that present these groups at the interface; alterna-tively, groups can be introduced onto the surface afterthe SAM is formed. Figure 2 gives examples of therange of groups that have been conjugated to SAMs.The properties of SAMs can be controlled further byformation of ‘mixed’ SAMs from solutions of two ormore alkanethiols. SAMs on gold are stable in air or incontact with water for periods of months. The mono-layers do undergo desorption at temperatures greaterthan 70 °C or when irradiated with ultraviolet (UV)light in the presence of oxygen. SAMs have sufficientstability in aqueous media for use in cell culture forperiods of days. The primary disadvantage encounteredin working with these monolayers is the requirement forgold-coated substrates. These substrates can be pre-pared by electron beam evaporation or sputtering of themetal, but they are not yet commercially available.
Monolayers that resist the adsorption of protein
The finding by Prime and Whitesides that monolayerspresenting short oligomers of the ethylene glycol (–OCH2CH2 –, EG) group are very effective at resistingthe adsorption of protein was critical to making thissurface chemistry broadly useful for biological applica-tions [25, 26]. Monolayers presenting either the shorttri(ethylene glycol) group or the longer hexa(ethyleneglycol) group were completely resistant to the adsorp-tion of protein. The degree to which these monolayerswere inert could be estimated by diluting the ethyleneglycol chains with a methyl-terminated alkanethiolate[–S(CH2)10CH3]. Studies using surface plasmon reso-nance (SPR) – a technique that measures the adsorp-tion of protein to interfaces in real time and in situ [27]– showed that SAMs wherein greater than 50% of thealkanethiolate chains present the glycol group resistedthe adsorption of virtually all proteins under a range ofsolution conditions; they even prevented the adsorptionof the ‘sticky’ protein fibrinogen [28]. The mechanismsby which these thin films resist adsorption are not yet
Figure 2. Representation of the structure of a SAM of alkanethi-olates on gold. The sulphur atoms coordinate to the gold and thetrans-extended alkyl chains present the terminal groups (X, Y) atthe interface. The table at the right gives examples (with refer-ences) of the variety of groups that have been attached to mono-layers.
M. Mrksich Tailored substrates for studies of attached cell culture656
well understood, but the use of these monolayers asinert interfaces has been critical to a number ofapplications (see below) [29] .
Methods for patterning the formation of monolayers
Substrates that are patterned into regions that alter-nately support or resist the attachment and spreading ofcells provide an effective method for examining theeffects of cell shape and form on cellular behaviour.Several groups have demonstrated methods forpreparing patterned substrates. These methods are allrelated in that they begin with a step that defines apattern on the substrate and then selectively modify thesurface to render selected regions inert. The mostimportant methods for patterning interfaces rely onphotolithography and microcontact printing.Photolithography is a technique that was developed forfabricating microelectronics circuits. The methodilluminates a substrate with UV light that is passedthrough a mask that has a pattern defined bytransparent and opaque regions. Upon exposure to thelight source, the substrate can be modified in a numberof ways. For SAMs of alkanethiolates on gold, the UVlight causes oxidative removal of the monolayer. Thepattern of exposed gold that is created can then bemodified with a monolayer presenting other groups [30].Photolithography has been much more important forpatterning monolayers of alkylsiloxanes on the surfacesof glass and silicon oxide. The common method startswith a silicon substrate coated with a thin layer ofphotoresist – a polymer that is degraded by UV light.Illumination of the substrate through mask, followed bya washing step, removes the polymer in exposed regionsto reveal a silicon oxide surface. A monolayer can thenbe assembled on these regions [31]. The remainingphotoresist is then removed by washing, and a differentalkylsiloxane can be formed in the complementaryregions. These techniques are very well developed. Theydo have the disadvantage, however, that thephotolithographic equipment and a controlledenvironment facility make these techniques expensive,and substantially less convenient than microcontactprinting.Microcontact printing (mCP) uses a rubber stamp toprint a patterned monolayer of alkanethiolates onto agold substrate [32]. The procedure is illustrated in figure3, and starts with the fabrication of a stamp usingphotolithography. A polished silicon wafer is coatedwith a thin layer of photoresist (a) and then exposed tointense UV light through a mask. Washing the substrateremoves exposed regions of photoresist (b). Theelastomeric stamp used in mCP is prepared by castingpolydimethylsiloxane (PDMS) against the patterned
Figure 3. A schematic illustration of mCP for patterning monolay-ers on gold. Each step of the process is discussed in the text.
CMLS, Cell. Mol. Life Sci. Vol. 54, 1998 657Review Article
photoresist (c). The PDMS stamp (d) is inked with asolution of the alkanethiol in ethanol (e), dried andmanually brought into contact with a gold surface (f);the alkanethiol is transferred only at those regionswhere the stamp contacts the surface (g). Conformalcontact between the elastomeric stamp and surface andthe rapid reaction of alkanethiols with gold permit thesurface to be patterned over areas several square cen-timetres in size with edge resolution of the featuresbetter than 100 nm. The regions of gold that remainafter the printing step can then be derivatized with adifferent SAM by immersing the substrate in a solutionof a second alkanethiol (h). The least convenient stepinvolves the photolithographic fabrication of the mas-ter pattern from which stamps are cast. Once prepared,multiple stamps can be cast from a single master andeach stamp can be used hundreds of times. mCP hasalso been used to pattern alkylsiloxanes on the surfacesof SiO2 and glass, but this method is not yet developedto the point of practical utility [33].
Studies of cell growth on patterned substrates
The combination of mCP and SAMs of alkanethiolateson gold provides a convenient and flexible methodol-ogy for controlling the positions and shapes of cellsattached to substrates [34]. The method uses mCP toprint a pattern of hydrophobic SAM followed by im-mersion of the substrate in a solution of oligo(ethyleneglycol)-terminated alkanethiol to render the nonprintedregions inert to protein adsorption and cell attachment.To insure efficient attachment of cells, these patternedsubstrates are immersed in a solution of matrix protein.Protein only adsorbs to the methyl-terminated regions:the oligo(ethylene glycol)-terminated regions resist en-tirely the attachment of cells [34]. This methodologycan create substrates that control the shape – andhence the growth – of individual cells [35]. We haverecently used this methodology to design substrates todetermine whether apoptosis of adherent endothelialcells is prevented by the total area of contact betweenthe cell and substrate or by the shape of the spread cell[36]. Figure 5 shows an example of a substrate pat-terned into several circles only a few microns in diame-ter: because the endothelial cells attached to several ofthese adhesive patches (but could not form adhesionsto areas between the patches), the projected area of thecell differed from the area of adhesion. A systematicstudy of adhesion on several substrates showed that cellshape – and not the total area of underlying matrix –was the important determinant of apoptosis [36]. Thismethodology has also been used to control the attach-ment of endothelial cells to surfaces contoured intogrooves and ridges [37]. I believe this methodology iscurrently the most flexible for controlling the interac-
Figure 4. Adhesion of endothelial cells on monolayers patternedinto regions terminated in methyl and hexa(ethylene glycol)groups. (A) Diagram of the patterns of monolayer used to controlthe shape and the total area over which cells contact matrix. (B)For sizes of adhesive circles with diameters greater than 20 mm,cells attached to a single patch and spread to the size of the patch;for smaller sized circles, cells attached to and spread on severalpatches [36].
tions of cells with substrates and that it will becomeincreasingly important as a tool in studies of cell-ma-trix interactions.The use of patterned alkylsiloxanes on glass substrateshas provided another important method for controllingcell adhesion [31, 38, 39]. Rudolph and co-workers, forexample, observed that endothelial cells allowed togrow on lines having a width of 100 mm differentiatedinto neovascular cords after a period of several days;cells that were seeded on lines having widths of 200 or500 mm, by contrast, had a less organized cytoskeletalstructure [38].
Monolayers that present ligands
SAMs can be tailored with ligands, peptides andproteins. The design of substrates that present ligandsfor the biospecific recognition of receptors must at thesame time prevent the nonspecific binding of otherproteins. If proteins do adsorb to the substrate, theadsorbed protein can both mask the immobilized lig-and and present other ligands from its primary se-quence.We have demonstrated that monolayers presenting oli-go(ethylene glycol) groups and ligands are very effec-tive for the biospecific adsorption of protein. As a first
M. Mrksich Tailored substrates for studies of attached cell culture658
system we investigated the binding of carbonic anhy-drase to SAMs presenting the benzenesulphonamidegroup and tri(ethylene glycol) groups [18]. Our workused SPR as the analytical technique to measure ad-sorption. The ability of this technique to measure bothkinetic and thermodynamic parameters for the associa-tion of soluble receptors with immobilized ligands, to-gether with a commercial source for the instrument,makes it very well suited for these studies. SPR showedthat carbonic anhydrase bound reversibly to thesemonolayers and that the amount of protein that boundwas proportional to the density of benzene-sulphonamide ligand in the SAM. Further, the adsorp-tion was shown to be biospecific, since the addition of asoluble, competitive ligand to the buffer inhibited bind-ing. This monolayer resisted the nonspecific adsorptionof protein when presented with a solution containingnine different proteins at a total concentration of 2mg/ml. Whitesides and coworkers have used a similarapproach to study the recognition of immobilized D-Ala-D-Ala dipeptides by the antibiotic vancomycin [21].This same strategy can be used to immobilize proteinsto the monolayers. Sigal and co-workers used a mono-layer that presented an Ni+2-nitrilotriacetic acid com-plex and tri(ethylene glycol) groups to immobilizeproteins whose primary sequence terminated in anoligo-histidine sequence [40]. This strategy has the ad-vantages that the his-tagged proteins can be immobi-lized from impure samples (provided there are no otherhis-tagged proteins present), and the immobilizedprotein is presented in a uniform orientation. Hong andco-workers described a related strategy for presentingproteins in a uniform orientation. These authors used amonolayer terminated in the thiol group to immobilizea cytochrome c through a disulphide linkage [23]. The
use of genetic engineering methods to generate mutantsof the protein having a single cysteine residue on thesurface allows the protein to be presented in a discreteorientation. The ease with which specific functionalgroups can be introduced into SAMs makes these inter-faces compatible with nearly all immobilizationchemistries (for a review of immobilization strategies,see ref. 41).
Functionalized substrates for cell adhesion
The finding by Ruoslahti and Pierschbacher that a keyinteraction in the adhesion cells on the fibronectin ma-trix involved binding of cellular integrin receptors toArg-Gly-Asp (RGD) peptides made possible a class ofsimple substrates for cell adhesion [42]. Early work usedpolymers that were derivatized with RGD – or longerpeptides that contain this sequence – to promote theadhesion of fibroblasts [43, 44]. In one example, poly-mer hydrogels modified with the GRGDS peptide sup-ported the morphologically complete spreading offoreskin fibroblasts [43]. These examples provide amodel system for understanding the relationships be-tween the density and structure of immobilized ligandswith the adhesion and spreading of cells. The use ofpolymer substrates, however, has the limitation that theenvironment of immobilized peptide is heterogeneous.Because not all of the peptides are accessible to cellularreceptors – they may be buried in the polymer – it isdifficult to control the density and homogeneity inbinding strength of immobilized ligands. The regularstructure of organic monolayers, by contrast, makesthese substrates an excellent choice for mechanisticstudies of cell adhesion.
Figure 5. Monolayers for the biospecific adsorption of protein. (Left) General structure of a monolayer presenting a ligand andtri(ethylene glycol) groups. (Right) The ligand-receptor combinations that have been demonstrated with this system include binding ofcarbonic anhydrase to benzenesulphonamide; his-tagged proteins to a complex of Ni+2; vancomycin to D-Ala-D-Ala.
CMLS, Cell. Mol. Life Sci. Vol. 54, 1998 659Review Article
Figure 6. (Left) Structure of a monolayer presenting the Arg-Gly-Asp peptide ligand and tri(ethylene glycol) groups. (Right) Opticalmicrographs of 3T3 fibroblasts attached to monolayers presenting peptide ligand at densities of 0.1 and 0.5% (after a period of 5 hours).
Massia and Hubbell prepared alkylsiloxane monolayersthat presented the RGD ligand in a homogeneous envi-ronment as substrates for the adhesion of human fore-skin fibroblasts [45]. These authors found a strongdependence of adhesion on the density of peptide. Atsurface densities of peptide equal to 1 fmol/cm2, thefibroblasts attached to and spread on the substrates, butdid not form stress fibres and focal adhesions. At a10-fold higher density of peptide, cells did recruit thesestructures. It was difficult to correlate the behaviour ofthese cells over longer periods of time because of thepossibility that the cells could remodel the matrix – thatis, that they could replace the underlying matrixproteins with secreted matrix proteins.We have used SAMs of alkanethiolates on gold thatpresent the GRGD peptide ligands and oligo(ethyleneglycol) groups as substrates with which to study celladhesion (fig. 6) [46]. For a series of monolayers havingdensities of RGD peptide decreasing from 1 to 0.001%,the number of attached endothelial cells decreased andthe degree to which the cells spread varied. Fluores-cence immunostaining showed that attached fibroblastsformed focal adhesions and stress fibres (fig. 7) (B. T.Houseman and M. Mrksich, unpublished observations).Two experiments suggest that while these monolayersallow cells to attach, they resist the deposition of matrixproteins by the cells. SPR showed that monolayers
presenting the peptide ligand at a density of 0.5% re-sisted the adsorption of several proteins [46]. Further,when cells were cultured in the presence of radiolabelledamino acids, the amount of matrix deposited on thesubstrate was substantially reduced compared with thaton conventional substrates that had adsorbedfibronectin.
Dynamic substrates
This review has described an important strategy forcreating substrates that have well-defined structures.The extension of this methodology to create substrateswhose structures and properties are under dynamic con-trol – for example, the ability to release adsorbedproteins or to change the presentation of ligands – willprovide an important method for further studies. Anumber of early examples have been described. Al-though these methods are not yet developed to thepoint of practical utility, they do provide a preview ofthe types of tailored substrates that will be developedover the next several years.Okano and co-workers used poly(N-isopropylacry-lamide) gels grafted to polystyrene dishes as substratesfor the adhesion of endothelial cells and hepatocytes[47]. These gels undergo a phase transition at tempera-
M. Mrksich Tailored substrates for studies of attached cell culture660
tures below 20 °C to give a material that resists theadsorption of protein. Consequently, when the culturedishes were removed from an incubator, the adherentcells were released from the dishes in a period of 30 min.We have developed a strategy that can selectively re-lease individual ligands from a monolayer of alkanethi-olates on gold [48]. The method attaches the ligand tothe monolayer through a redox-active group that under-goes electrochemical oxidation (using the gold as work-ing electrode) and subsequent cleavage to release theattached ligand. Because the monolayer is stable to the
applied electrical potentials, it will be possible to createsubstrates that present multiple groups but selectivelyrelease only a fraction of these groups.Electrical fields present at the interface of a conductingsubstrate can affect the behaviour of cells. Langer andco-workers used electrically conducting polypyrrolefilms as substrates for studies of neurite outgrowth inPC-12 cells [49]. Cultured cells subjected to an electricalstimulus produced neurite lengths that were greater by afactor of 2 relative to those cultured in the absence offields. A related study showed that reversible oxidationof these polymer films could stall the growth (includingcell extension and DNA synthesis) of aortic endothelialcells [50]. Dynamic fields generated by applying an ACcurrent to regions of a substrate have been shown torepel cells in suspension and prevent adhesion [51]. Themechanisms by which applied fields can influence thebehaviour of cells are not yet well understood, but theability to control both the spatial localization and inten-sity of fields offers many opportunities for creatingfunctional substrates.Techniques common to silicon microfabrication provideaccess to a range of tailored substrates that can measuremechanical and physiological properties of cells [52].Galbraith and Sheetz, for example, used micromachin-ing to create a substrate that had several thousandmicron-sized pads that could measure forces exerted byan attached cell [53]. They used this substrate to mea-sure the distribution of traction forces exerted on asubstrate by a migrating fibroblast [53]. In a programmeto develop cell-based sensors, Kovacs has fabricatedelectrode arrays for measuring the response of attachedneural cells to chemical toxins [54].SAMs of alkanethiolates on gold are currently the bestclass of substrates that can control the structure, densityand pattern of immobilized ligands. The ease withwhich these substrates can be prepared, the syntheticflexibility available in attaching different groups, andthe compatibility with the conditions and techniques ofcell culture make this methodology broadly useful forstudies of attached cell culture. This review highlightedseveral examples that utilized tailored substrates forstudies in experimental cell biology. The continued col-laborations of biologists, chemists and surface scientistswill indeed lead to many more examples.
Acknowledgements. For support of research in my laboratory, Iam grateful to the National Institutes of Health, the NationalScience Foundation, the Searle Scholars Program/The ChicagoCommunity Trust, the Camille and Henry Dreyfus Foundation,the Army Materiel Command and DARPA.
1 Ramsden J. J. (1995) Puzzles and paradoxes in protein adsorp-tion. Chem. Soc. Rev. 73–78
2 Wahlgren M. and Arenbrant T. (1991) Protein adsorption tosolid surfaces. Trends Biotechnol. 9: 201–208
Figure 7. Fluorescence immunostaining of F-actin (A) and vin-culin (B) in a fibroblast attached to a monolayer presenting 1%RGD peptide as illustrated in figure 6. Cells were fixed for 30 minin 4% paraformaldehyde/phosphate-buffered saline and permeabi-lized with 0.1% Triton X-100. Labelling was performed usingrhodamine-conjugated phalloidin and a fluorescein immunoglobu-lin G conjugate to monoclonal antivinculin.
CMLS, Cell. Mol. Life Sci. Vol. 54, 1998 661Review Article
3 Sadana A. (1992) Protein adsorption and inactivation onsurfaces, influence of heterogeneities. Chem. Rev. 92: 1799–1818
4 Lee C.-S. and Belfort G. (1989) Changing activity of ribonucle-ase A during adsorption: a molecular explanation. Proc. Natl.Acad. Sci. USA 86: 8392–8396
5 Steadman B. L., Thompson K. C., Middaugh C. R., MatsunoK., Vrona S., Lawson E. Q. et al. (1992) The effects of surfaceadsorption on the thermal stability of proteins. Biotechnol.Bioeng. 40: 8–15
6 Ramsden J. J., Roush D. J., Gill D. S., Kurrat R. and WillsonR. C. (1995) Protein adsorption kinetics drastically altered byrepositioning a single charge. J. Am. Chem. Soc. 117: 8511–8516
7 Kull T., Nylander T., Tiberg F. and Wahlgren N. M. (1997)Effect of surface properties and added electrolyte on thestructure of b-casein layers adsorbed at the solid/aqueousinterface. Langmuir 13: 5141–5147
8 Siegel R. R., Harder P., Dahint M., Grunze M., Josse F.,Mrksich M. et al. (1997) On-line detection of non-specificprotein adsorption at artificial surfaces. Anal. Chem. 69: 3321–3328
9 Mrksich M. (1997) Using self-assembled monolayers to under-stand the biomaterials interface. Curr. Op. Coll. Interface Sci.2: 83–88
10 Mrksich M. and Whitesides G. M. (1996) Using self-assembledmonolayers to understand the interactions of man-made sur-faces with proteins and cells. Annu. Rev. Biophys. Biomol.Struct. 25: 55–78
11 Hanein D., Sabanay H., Addadi L. and Geiger B. (1993)Selective interactions of cells with crystal surfaces. J. Cell Sci.104: 275–288
12 Boyan B. D., Batzer R., Kieswetter K., Liu Y., Cochran D. L.,Moncler S.-S. et al. (1998) Titanium surface roughness altersresponsiveness of mg63 osteoblast-like cells to 1a,25-(OH)2D3.J. Biomed. Mat. Res. 39: 77–85
13 Plant A. L., Brigham-Burke M., Petrella E. C. and O’ShannessyD. J. (1995) Phospholipid alkanethiol bilayers for cell-surfacereceptor studies by surface-plasmon resonance. Anal. Biochem.226: 342–348
14 Ulman A. (1996) Formation and structure of self-assembledmonolayers. Chem. Rev. 96: 1533–1554
15 Dubois L. H. and Nuzzo R. G. (1992) Synthesis, structure andproperties of model organic surfaces. Annu. Rev. Phys. Chem.43: 437–463
16 Bain C. D. and Whitesides G. M. (1989) Modeling organicsurfaces with self-assembled monolayers. Adv. Mat. in Angew.Chem. 101: 522–528
17 Motesharei K. and Myles D. C. (1994) Molecular recognitionin membrane mimics: a flourescence probe. J. Am. Chem. Soc.116: 7413–7414
18 Mrksich M., Grunwell J. R. and Whitesides G. M. (1995)Bio-specific adsorption of carbonic anhydrase to self-assembledmonolayers of alkanethiolates that present benzenesulfonamidegroups on gold. J. Am. Chem. Soc. 117: 12009–12010
19 Spinke J., Liley M., Guder H.-J., Angermaier L. and Knoll W.(1993) Molecular recognition at self-assembled monolayers: theconstruction of multicomponent multilayers. Langmuir 9:1821–1825
20 Whitesell J. K., Hyung H. and Whitesell C. S. (1994) Enzymaticsmoothing of thin organic layers. Angew. Chem. Int. Ed. Engl.33: 871–873
21 Yan L., Rao J. and Whitesides G. M. Using surface plasmonresonance to study the binding of vancomycin and its dimer toself-assembled monoalyers presenting D-ala-D-ala. J. Am.Chem. Soc., in press
22 Delamarche E., Sundarababu G., Biebuyck H., Michel B.,Gerber Ch., Sigrist H. et al. (1996) Immobilization of antibodieson a photoactive self-assembled monolayer on gold. Langmuir12: 1997–2006
23 Hong H. G., Jiang M., Sligar S. G. and Bohn P. W. (1994)Cysteine-specific surface tethering of genetically engineeredcytochromes for fabrication of metalloprotein nanostructures.Langmuir 10: 153–158
24 Chrisey L. A., O’Ferrall C. E., Spargo B. J., Dulcey C. S. andCalvert J. M. (1996) Fabrication of patterned DNA surfaces.Nucleic Acids Res. 24: 3040–3047
25 Prime K. L. and Whitesides G. M. (1991) Self-assembledorganic monolayers: model systems for studying adsorption ofproteins at surfaces. Science 252: 1164–1167
26 Prime K. L. and Whitesides G. M. (1993) Adsorption ofproteins onto surfaces containing end-attached oligo(ethyleneoxide): a model system using self-assembled monolayers. J. Am.Chem. Soc. 115: 10714–10721
27 Silin V. and Plant A. (1997) Biotechnological applications ofsurface plasmon resonance. Trends Biotechnol. 15: 353–359
28 Mrksich M., Sigal G. B. and Whitesides G. M. (1995) Surfaceplasmon resonance permits in situ measurement of proteinadsorption on self-assembled monolayers of alkanethiolates ongold. Langmuir 11: 4383–4385
29 Mrksich M. and Whitesides G. M. (1997) Using self-assembledmonolayers that present oligo(ethylene glycol) groups to con-trol the interactions of proteins with surfaces. In: Poly(ethyleneglycol): Chemistry and Biological Applications, vol. 680, pp.361–373, ACS Symposium Series, Washington, DC
30 Tarlov M. J., Burgess D. R. F. Jr. and Gillen G. (1993) UVphotopatterning of alkanethiolate monolayers self-assembledon gold and silver. J. Am. Chem. Soc. 115: 5305–5306
31 Kleinfeld D., Kahler K. H. and Hockberger P. E. (1988)Controlled outgrowth of dissociated neurons on patternedsubstrates. J. Neurosci. 8: 4098–4120
32 Mrksich M. and Whitesides G. M. (1995) Patterning self-assem-bled monolayers using microcontact printing: a new technologyfor biosensors? Trends Biotechnol. 13: 228–235
33 Xia Y., Mrksich M., Kim E. and Whitesides G. M. (1995)Patterning siloxane monolayers on the surfaces of silicondioxide with microcontact printing. J. Am. Chem. Soc. 117:9576–9577
34 Mrksich M., Dike L. E., Tien J. Y., Ingber D. E. and WhitesidesG. M. (1997) Using microcontact printing to pattern theattachment of mammalian cells to self-assembled monolayers ofalkanethiolates on transparent films of gold and silver. Exp.Cell Res. 235: 305–313
35 Singhvi R., Kumar A., Lopez G. P., Stephanopoulos G. N.,Wang D. I. C., Ingber D. E. et al. (1994) Engineering cell shapeand function. Science 264: 696–698
36 Chen C. S., Mrksich M., Huang S., Whitesides G. M. andIngber D. E. (1997) Geometric control of cell life and death.Science 276: 1425–1428
37 Mrksich M., Chen C. S., Xia Y., Dike L. E., Ingber D. E. andWhitesides G. M. (1996) Controlling cell attachment on con-toured surfaces with self-assembled monolayers of alkanethio-lates on gold. Proc. Natl. Acad. Sci. USA 93: 10775–10778
38 Spargo B. J., Testoff M. A., Nielsen T. B., Stenger D. A.,Hickman J. J. and Rudolph A. S. (1994) Spatially controlledadhesion, spreading and differentiation of endothelial cells onself-assembled monolayers. Proc. Natl. Acad. Sci. USA 91:11070–11074
39 Britland S., Clark P., Connolly P. and Moores G. (1992)Micropatterned substratum adhesiveness: a model for morpho-genetic cues controlling cell behavior. Exp. Cell Res. 198:124–129
40 Sigal G. B., Bamdad C., Barberis A., Strominger J. andWhitesides G. M. (1996) A self-assembled monolayer for thebinding and study of histidine-tagged proteins by surfaceplasmon resonance. Anal. Chem. 68: 490–497
41 Mosbach K (1987) Immobilization of proteins and cells. Meth-ods Enzymol. 135: 1–675
42 Ruoslahti E. and Pierschbacher M. D. (1987) New perspectivesin cell adhesion: rgd and integrins. Science 238: 491–497
43 Brandley B. K. and Schnaar R. L. (1988) Covalent atttachmentof arg-gly-asp sequence peptide to derivatizable polyacrylamidesurfaces: support of fibroblast adhesion and long-term growth.Anal. Biochem. 172: 270–278
44 Drumheller P. D. and Hubbell J. A. (1994) Polymer networkswith grafted cell adhesion peptides for highly biospecific celladhesive substrates. Anal. Biochem. 222: 380–388
M. Mrksich Tailored substrates for studies of attached cell culture662
45 Massia S. P. and Hubbell J. A. (1991) An rgd spacing of 440nm is sufficient for integrin avb3-mediated fibroblast spreadingand 140 nm for focal contact and stress fiber formation. J. Cell.Biol. 114: 1089–1100
46 Roberts C., Chen C. S., Mrksich M., Martichonok V., IngberD. E. and Whitesides G. M. Using mixed self-assembledmonolayers presenting GRGD and EG3OH groups to charac-terize long-term attachment of bovine capillary endothelial cellsto surfaces. J. Am. Chem. Soc. in press
47 Okano T., Yamada N., Okuhara M., Sakai H. and Sakurai Y.(1995) Mechanism of cell detachment from temperature-modu-lated, hydrophilic-hydrophobic polymer surfaces. Biomaterials16: 297–303
48 Hodneland C. H. and Mrksich M. (1997) Design of self-assem-bled monolayers that release attached groups using appliedelectrical potentials. Langmuir 13: 6001–6003 49 Schmidt C. E.,Shastri V. R., Vacanti J. P. and Langer R. (1997) Stimulationof neurite outgrowth using an electrically conducting polymerfilm. Proc. Natl. Acad. Sci. USA 94: 8948–8953
50 Wong J. Y., Langer R. and Ingber D. E. (1994) Electricallyconducting polymers can noninvasively control the shape andgrowth of mammalian cells. Proc. Natl. Acad. Sci. USA 91:3201–3204
51 Schnelle T., Muller T., Voigt A., Reimer K., Wagner B. and FuhrG. (1996) Adhesion-inhibited surfaces. Coated and uncoatedinterdigitated electrode arrays in the micrometer and submicrom-eter range. Langmuir 12: 801–809
52 Kovacs G. T. A., Petersen K. and Albin M. (1996) Siliconmicromachining: sensors to systems. Anal. Chem. 68: 407A-412A
53 Galbraith C. G. and Sheetz M. P. (1997) A micromachined deviceprovides a new bend on fibroblast traction forces. Proc. Natl.Acad. Sci. USA 94: 9114–9118
54 Borkholder D. A., Opris I. E., Maluf N. I. and Kovacs G. T.A. (1996) Planar electrode array systems for neural recordingand impedance measurements, Conference Proceedings of theIEEE, Engineering in Medicine and Biology, Amsterdam, TheNetherlands
![University of Groningen Enzymatic Glycosylation of Small … · 2016. 3. 5. · Enzymatic Glycosylation of Small Molecules: Challenging Substrates Require Tailored Catalysts Tom Desmet,[b]](https://static.documents.pub/doc/80x56/60c2f6dd9afff146c04bf734/university-of-groningen-enzymatic-glycosylation-of-small-2016-3-5-enzymatic.jpg)
